MLOps Engineer

Role OverviewWe are looking for an experienced MLOps Engineer to build and scale our AI infrastructure across Kubernetes, cloud-native environments, and serverless GPU platforms. You will own the end-to-end operational lifecycle of machine learning models—from training to deployment, monitoring, optimization, and automated retraining.Key Roles

• Design and implement highly scalable AI / ML infrastructure using Kubernetes, Kubeflow, Ray, and cloud-native services.
• Build robust CI / CD and CT (Continuous Training) pipelines for model deployment, inference, monitoring, and automated retraining.
• Architect and deploy ML workflows on serverless GPU platforms (AWS, GCP, Yotta, RunPod, Modal, etc.) for cost-efficient, elastic scaling.
• Establish automated systems for model drift, data drift, performance monitoring, and lineage tracking.
• Promote best practices in reproducible ML, infrastructure-as-code, automation, and internal tooling.Responsibilities
• Evaluate, integrate, and optimize MLOps tools (MLflow, Weights & Biases, KServe, Seldon, BentoML, Argo, Airflow, etc.) to streamline AI development.
• Develop scalable inference-serving layers—batch, real-time, streaming—using GPU-optimized serving frameworks.
• Build observability stacks for GPU utilization, latency, throughput, and model health metrics.
• Implement robust systems for model governance, versioning, rollout strategies (blue / green, canary), and automated rollback.
• Collaborate closely with ML engineers, data engineers, and product teams to deliver production-ready AI features.Knowledge & Skills Requirements
• Strong understanding of ML / DL fundamentals and hands-on experience with model training and optimization.
• Expertise in Kubernetes, containerization, Helm, and cloud-native infrastructure.
• Experience with serverless GPU architectures and distributed computing frameworks.
• Solid knowledge of CI / CD tools (GitHub Actions, GitLab CI, Jenkins), IaC (Terraform), and workflow engines.
• Understanding of drift detection, performance tracking, experiment management, and scalable model deployment patterns.We Accept International Applicants